1. Field of the Invention
The present invention relates to a data read/write device having high recording density.
2. Description of the Related Art
In recent years, a small sized portable device has been prevalent worldwide. At the same time, with significant progress of a high-speed data transmission network, a demand for a small sized large capacity nonvolatile memory has rapidly increased. Among them, a NAND-type flash memory and a small sized hard disk drive (HDD) have made rapid progress in recording density, and a large market has been formed.
However, in both of them, a limit of recording density has already been pointed out. That is, there is a problem that an increase of processing cost caused by reduction of a minimum line width becomes remarkable in the NAND-type flash memory and tracking precision cannot be sufficiently allocated in a small sized HDD.
There have been proposed some ideas of novel memories aiming to significantly exceed the limit of recording density, under such a situation.
For example, a phase change memory (PRAM) uses a recording material capable of taking two states, i.e., an amorphous state (ON) and a crystalline state (OFF). A principle of recording data is that the two states are associated with binary data “0” and “1”, respectively.
With respect to a write/erasure operation, for example, the amorphous state is produced by applying a large electric power pulse to a recording material while the crystalline state is produced by applying a small electric power pulse to a recording material.
A read operation is made by supplying a small amount of a read current to a recording material to such an extent that write/erasure does not occur, and then, measuring an electrical resistance of the recording material. A resistance value of the recording material in the amorphous state is greater than that of the recording material in the crystalline state, and a difference therebetween is in order of 103.
The maximum feature of the PRAM is that operation can be made even if an element size is reduced to an order of 10 nm. In this case, the recording density of about 1.5 Tbpsi (terra bite per square inch) can be achieved, thus providing one of the candidates for the achievement of high density recording (refer to, for example, JP-A 2005-252068(KOKAI)).
Although different from the PRAM, there has been reported a novel memory having a principle of operation that is very similar to the PRAM (refer to, for example, JP-A 2004-234707(KOKAI)).
According to this report, a typical example of the recording material for recording data is nickel oxide. Like the PRAM, a large electric power pulse and a small electric power pulse are used for a write/erasure operation. In this case, there has been reported an advantage that power consumption at the time of the write/erasure operation is reduced as compared with the PRAM.
Although, up to now, an operational mechanism of this novel memory has not been clarified, its reproducibility is verified, thus providing another one of the candidates for the achievement of high density recording. In addition, with respect to the operational mechanism as well, some groups have attempted to clarify the mechanism.
In addition to these memories, an MEMS memory using a MEMS (micro electro mechanical systems) technique has been proposed (refer to, for example, P. Vettiger, G. Cross, M. Despont, U. Drechsler, U. Durig, B. Gotsmann, W. Haberle, M. A. Lants, H. E. Rothuizen, R. Stutz and G. K. Binning, IEEE Trans. Nanotechnology 1, 39 (2002)).
In particular, a MEMS memory called Millipede has a structure in which a plurality of array-shaped cantilevers and recording media having an organic substance applied thereto are opposed to each other. A probe of a distal end of the cantilever comes into contact with the recording medium at a proper pressure.
The write operation is selectively made by controlling a temperature of a heater which is added to the probe. That is, if the heater temperature is increased, the recording medium is softened, the probe sinks into the recording medium, and then, a cavity is formed in the recording medium.
The read operation is made in such a manner that, while a current to such an extent that the recording medium is not softened is supplied to a probe, the probe is made to scan on the surface of the recording medium. If the probe falls into the cavity of the recording medium, the probe temperature decreases, and then, the resistance value of the heater increases. Thus, data can be sensed by reading a change of the resistance value.
The maximum feature of the MEMS memory such as Millipede is that the recording density can be remarkably improved because it is necessary to provide wiring at each recording portion for recording bit data. As it now stands, the recording density of about 1 Tbpsi has already been achieved (refer to, for example, P. Vettiger, T. Albrecht, M. Despont, U. Drechsler, U. Durig, B. Gotsmann, D. Jubin, W. Haberle, M. A. Lants, H. E. Rothuizen, R. Stutz, D. Wiesmann and G. K. Binnig, P. Bechtold, G. Cherubini, C. Hagleitneri, T. Loeliger, A. Panmtazi, H. Pozidis and E. Eleftheriou, in Technical Digest, IEDM03 pp. 763-766).
Upon the receipt of Millipede, recently there has been made an attempt to achieve remarkable improvement with respect to power consumption, recording density, an operating speed and the like by combining a MEMS technique and a new principle of recording.
For example, there has been proposed a system of providing a ferroelectric layer at a recording medium, and then, applying a voltage to the recording medium, thereby inducing dielectric polarization in the ferroelectric layer to record data. According to this system, there is a theoretical prediction that a gap (recording minimum unit) between recording portions for recording bit data can be approached to a unit bulla level of a crystal.
Assuming that a minimum unit of recording becomes 1 unit bulla of the crystal of the ferroelectric layer, the recording density is obtained as a very large value of about 4 Pbsi (pico bite per square inch).
However, even up to now, such a MEMS memory capable of ferroelectric recording has not been achieved, although it is a conventionally known principle.
The largest reason is that an electric field coming out of the recording medium to the outside thereof is interrupted by ions in air. Namely, the electric field from the recording medium cannot be sensed, thus disabling a read operation.
There is another reason that, when a lattice defect exists in a crystal, an electric charge caused by such a lattice defect moves to a recording portion, interrupting the electric charge.
The former problem with electric field interruption caused by the ions in the air is solved by proposing a read system using a scanning type nonlinear dielectric microscope (SNDM), and this novel memory is remarkably progressed for the achievement of practical use (refer to, for example, A. Onoue, S. Hashimoto, Y. Chu, Mat. Sol. Eng. B 120, 130 (2005)).